XeKr

Nope. Really great for Lunatic+ too. I may end up taking more turns to satisfy my reliability cravings, so hopefully there's enough exp. >_>

Chrom only got to 12/1 in Redwall's run and his Str looks blessed too (21 Str at 12/1 where 11 is the personal+Great Lord base. There may be statboosters in there.). I'm unsure how many more levels is feasible without slowing down by a bunch. Anyways, I might have a bit more exp than his run, but probably not by much (optional paralogues).

While this is “efficiency”, I favor a more stringent metric (expected turn count) than other runs, and so I still currently anticipate 1-2 turning many later maps, as movement, Rescue spam, and reliability allows. As such, at this point I don’t think there’s enough exp to train both Morgan and Lucina to that extent (from what I understand, you’re suggesting Lucina in front with a maxed Morgan as her support). I expect Chrom to barely get any levels, as well. But that calculation is pretty representative of what I’m anticipating with Avatar + Lucina. Avatar chips a bit of damage with a forged Brave, Lucina doing maybe 15ish per Dual Strike (those don't get halved by Pavise). Panne might be there to chip a bit more. It’s also pretty hard to double Grima without a +Spd Pair Up, as even the Spd cap + tonic + Rally Speed often isn’t enough. Otherwise +Str Pairs would definitely be good for more damage. Though Great Lord Lucina can maybe do it. Of course it’s possible to rig Aether or Luna as a failsafe, but I’d prefer to avoid that if possible. Though it might turn out something silly like Lucina attacking with a forged Brave and not her Falchion + 8 Aether/Luna chances + Hero Avatar as support with a forged Brave is more reliable. If Grima is not killed in 1 turn, however, he usually heals back to max. The reliable 2 turn is to get the necessary infrastructure (Anathema, Hex, Charm, etc) to improve the odds. If it's not done in 2 turns it's probably not getting done. >_>

That's strange, since they work when adjacent and such, so it's not like Pair Up. I suppose we'll see...

Lucina and Tiki forever. <3. I'm also partial to Lon'qu, Panne, Libra and a bunch of children like Nah, FeMorgan, Cynthia, Owain, Gerome.

Can get +10-20 hit from Support, which I think still applies even if she's in the back. Does Outdoor Fighter count? >_> But yes, certainly an issue.

Yeah, by that time Lucina's class doesn't really matter, so I might go Hero/Great Lord or something for -1ish damage and +7ish hit. I'll also definitely try to use Anathema/Hex to lower avoid a solid bit. Though I'm not sure if Grima's throne/Ire bonus is already included or not too... Regardless, Chrom is not getting anywhere near max stats, so he'd be worse off I think. I anticipate going through Hero for Sol after Griffon (that's also part of why Panne is there, so I still have a flier). That's a bit more Spd, and maybe I'll save Naga's Tear. 5% growth over a bunch of levels is a decent chunk as well. It depends on how a lot of things shake out. Though I think my Spd is already (barely) below average, xd. It's very possible that since +Def doesn't snowball anywhere near as fast as I expected, that +Spd is better for more reliability. I would love to have Rally Spectrum, but I don't see how to fit it in without other sacrifices.

I'm less concerned about the damage issue compared to the accuracy issue. You "only" have to hit for 12 damage 8 times + 3 more damage to kill. And maybe it's even more leniant if you count enemy phase (sacrificing various characters). Something like Celica's Gale would be ideal, but I'm going for Deliverer instead of Galeforce and not using Morgan, so there's not really any users. I think it's hard to damage Grima with the lead unit in general, anyway. Unless it's Luna Celica's or something. As for Lucina, I was planning on Lord->Cav->Great Knight. The theory is Aether/Luna makes offense more reliable. High Hp/Def bases + Aether + maybe Dual Guard+ ensures survivability. I am concerned about her stats, particularly Spd and such, however.

We shall see I guess. I think one of the more interesting things about FE13 Lunatic is there are noticeable tradeoffs to be done, where taking turns to train earlier can likely payoff in (much) turn shaving later on. It’s very drastic here because of Veteran, and quite hard to optimize. Also there’s tradeoff in terms of mobility vs. durability, where self-healing/super concrete durability may be necessary for reliable clears yet those classes are quite slow moving. I’m also finding it hard to get the snowball rolling since even +Def Avatar can’t really tank yet without slowing down a bunch (or at least a bit more than I’d like). I suppose small things like getting 3hko'd vs. 2hko'd do help. The final boss issue is definitely tricky as I don’t think you can train Chrom to be that strong in a reasonable timeframe (unlike FE12 Marth). He can do solid damage to the final boss somewhat reliably even without too many levels, but I’m going to try to see if Lucina can do it in my run. That’s why I’m forgoing Morgan as a combat unit, as I don’t think there’s enough exp to train both. And because Lucina is <3.

iirc in HM, once Donnel isn't doubled (after a level or 2), he can actually tank stuff with Kellam support and Tonics.

Are you sure it's true in Normal and Hard? I didn't find him hard to train at all my first run. Plus, Donnel comes around Chapter 4. Nino comes around Chapter 26/28. At least if I train Donnel for a chapter or 2 he demolishes the rest of the game. <_<

The prevalent opinion seems to be that he completely sucks, even though he’s exactly the type of unit many people like to train (in the vein of Nino, FE8 trainees). He’s quite good when trained too, unlike those units. He’ll outstat most characters for the majority of the game after getting over a tiny hump (in NM, HM). He has easy access to 100% Armshrift Helswath and Sol. Aptitude makes some kid amazing. He gives both Galeforce and Armshrift access to some lucky child if desired. Etcetc I do think he’s the “worst” character by a good margin, but I had lots of fun with Donnel and Donnel!Owain my very first FE13 run. >_>

Poor Donnel, simultaneously overrated and underrated by different player populations. ;_;

If it's your first (no grinding) Lunatic run, I would recommend Avatar + Nowi for Manakate daughters with Veteran as an alternative to the popular Chrom + Avatar pairing. It's also a nice pairing for postgame. But you should really marry for love.

Though this site has a few enemy/boss stats, imo it's hard to gauge that kind of stuff relative to your own characters until doing a run yourself. Especially since this game has so many "growth characters" and not really any fail-safes, as you say (Tiki kinda, but still requires some training). I do think a casual runthrough would help, though it's still very different. What really gets me is starting from around C13, it's pretty much an onslaught of bosskills until C23. It's going from unpromoted stuff to insane 30-40 stat enemies with +8 mt, +20 hit forges really fast. And it’s a rout with super-high enemy density. Another thing to note is other FE games don't really have a character that levels so insanely fast like Avatar or Morgan, especially considering the exp floor in this game. In other games being a couple levels above the enemy or being promoted quickly diminishes exp to tiny amounts like 4 exp, while in this game Avatar will still get like 15-20 exp despite being overleveled, and Morgan around recruitment time easily gets 100 exp per kill for a while. The snowball effect is pretty crazy. Buyable Rescue staves certainly don't help matters. >_> Anyways, here’s Chapter 2. [spoiler=Chapter 2: 7/18] So the strategy usually results in a 6 or 7 turn and I make a few decisions to make it less of a 6 turn or death. I just got a bit unlucky once everything was somewhat ironed out, but I’m fine with 7 turns. Though Level ups weren’t great either. :( Anyways, the only feasible way to clear efficiently is with Frederick. Even if Avatar can tank, her damage output is simply too low to clear quickly. So, the first decision is who Pairs with Frederick. Not Avatar, because we want her to feed her some Veteran kills. Not Sully, because her initial position is very important and she needs to save some of the other units first turn. Vaike is actually a decent option since with +Str levels, Frederick can one-shot Mercs. However, I think my Frederick does not have the Str, it’s unreliable, and moreover, Vaike has no weapon, so no Dual Strike chance (this is a little important as I’ll note later). It turns out Frederick shouldn’t pair with Chrom either, despite doubling Barbs with him and having the best Dual Strike chance from C support and Dual Strike+. Chrom needs a few of his own kills, and is one of the few who can take a hit reliably. Plus, he gives no Def, which can actually matter for Frederick’s survival. This leaves Stahl or Virion and I choose Virion since they support, he’s very accurate, and so Avatar can have Stahl’s Bronze Sword without influencing other stuff. In addition, Stahl's +1 Spd makes it easier for Avatar to not get doubled (my Avatar was solidly average in Spd, but others may not be). The first priority is to make some precise positioning to survive the first turn and set up a something that can deflect pressure for the subsequent turns. The most threatening initial enemies are the Barb and the Merc to the left. The Merc in particular doubles many characters so he must be dealt with first. Frederick with Virion hits him hard with the Silver Lance. Avatar trades for Stahl’s Bronze Sword, Pairs with him, and they go to finish the Merc if Virion did not Dual Strike (Even if it is dead, they move there, below Frederick). They also trade Frederick the Vulnerary and Elixer that Avatar has. It is here we see that Avatar only has ~65% or ~80% true hit depending on if the Merc has Patience. Vaike pairs with Chrom and they go left into a forest, Chrom in front. Sully and Lissa Pair and head left. Sully drops Lissa near the bottom so Miriel has a safe place to go when she arrives. If both the Dual Strike and Avatar’s attack missed, there’s a final contingency in Chrom. Chrom can counterkill that Merc, and may be able to dodge or Dual Guard an attack from the Merc or Barb. I’d estimate >90% chance of something good working. If the Merc is already dead, usually the Barb attacks Chrom or Avatar, it doesn’t matter. At this point, Frederick heads up, trying to stay in good terrain if possible. He uses the Silver Lance if there are tons of enemies around and heals with whatever item is more appropriate. Avatar heads down to help out Chrom and whoever is hurt is healed by Lissa Paired with Miriel. It seems that the enemy move order is pretty set. Therefore on enemy phase the Barb usually attacks Chrom which blocks him from the Soldiers so they suicide into Avatar instead. Even though the enemy could kill units if suiciding the Soldiers first, they don’t seem to think of that. I suppose it’s also possible to unequip characters if necessary for more consistency but I didn’t test it that rigorously. As Frederick continues moving up, he can unequip if he’s left some enemies at low health for Avatar and Chrom to pick off. Regardless, he should be fighting up top by turn 4. Turn 5, Frederick should typically be attacking the Boss with a sword (ideally an Iron Sword, so a Silver Lance hit finishes, regardless of terrain). If positioning works out so Frederick is on a fort or somehow has a bunch of Hp still, he can finish on Turn 6. Otherwise he needs to spend it healing to finish fairly consistently on Turn 7. I suppose there’s a tiny but non-trivial 8 turn chance, plus Frederick can get crit or something but that's always going to be there. Avatar gets whatever kills she can. There are probably various cases to feed more kills, including the boss, to Avatar or Chrom, but I think they’re fairly rare and I didn’t feel like reseting for them. So 7 turns, though 6 is a possibility with this strat. Exp values can vary wildly depending on how many Dual Strikes were triggered or various random positionings but I have level 2 Chrom and level 7 Avatar. I do wonder if it's worth taking a few more turns to snowball Avatar faster. Oh well... I’m sure there’s much more optimization to be done on this chapter, particularly if people are sacrificed, but I really just want to move on.

Gaius makes a nice Myrmidon to preserve his sword rank. It's particularly nice early on because you'll have Avoid+10 and triangle advantage against the many Axe users. Later on, transition into Hero for Sol and Hand Axes. Fighter can work too because he's so naturally fast, though as noted, a bit of a glass cannon. Avatar can be anything. If using a larger team, I like grabbing Rally Spectrum first, then go into any other magic class. The children issue depends on if you're planning for "min-maxing" or not. It's not necessary, but it can be fun to optimize certain class sets and cap modifiers. There are alternatively other usable pairings for the main game if you want to use children during that.

I understand as well. Just I’m of the opinion that this metric satisfyingly explains many of the intuitive conclusions that have been reached in previous tier lists. Such as why, in general, Jagens are good, Ests are bad, mounted/flying units dominate, Dancers are good, etc. How are you defining payoff here though? Donnel becomes an immortal 1-2 range one-rounding god, fairly quickly after he reclasses. Nowi just requires less turns to get there (maybe…) and she’s appropriately ranked much higher. If we’re then thinking in terms of turns, both still require a really large number compared to the Veteran characters. Okay, like I said, I don’t think we really disagree that much. I don’t think there’s any place itt where there’s nitpicking over a single turn or so (it’s something that’s typically covered by the reliability constraint). We aren’t saying various characters can never get that exp, we’re saying it requires substantially slowing down to get that exp. There have been concessions in terms of character roles in varying team compositions. I apologize if my earlier wording implied that Sully is “trash” or “useless”, just that I think she’s less efficient (requires much slower and/or less reliable clears) than many others.

None of those criteria say anything about what characters are better; there is no specified metric there. I feel it would reduce to an "efficiency"-type list (eventually, by way of questions like what does it mean to be "good" at routing, or getting chests?), just there's more secondary objectives to complete. Admittedly, it may mitigate distinctions between characters (arguable if it will, plus if that's good or not), but any number of arbitrary criteria can do that (i.e., must get all items, must get all talk convos, must not reclass to mounted classes, must deploy lowest level units, must gain x amount of exp per chapter, etcetc). I mean, in the end, tier lists are entirely dependent on interest in them, so I'd favor giving it a shot as well I suppose. EDIT: If you think about it, the present lists are primarily concerned with "completion." It's just that we value characters that complete the game quickly and reliably, i.e we consider that to be "good".

I think the subset of players that value this metric is quite small. Nice work, though. I think your solution is quite simple and intuitive, just no one (well, I didn’t. >_>) really bothered to sit down and do the math. Tiering is srs business. It may be helpful to define the conditional mean turncounts, for reference.

You’re very right. I only brought up that point to illustrate that players take those kinds of risks all the time. There is no 100% safe strategy. Obviously there are tons of other stuff at play when you start to discuss real-time efficiency. Honestly should just speedrun. >_>

I'm someone who greatly prefers minimal chance of death as well, when possible. However, it's completely impossible to get 100% strats, particularly when accounting for the RNG in growths. Do you reject a strategy that relies on dodging a 1% disp hit? 10%? Where's the cutoff of what you deem allowable and not? This is what expected turn count resolves.This is the purpose of using statistics to summarize a wide range of possibilities.

It seems unsatisfying to not be able to provide those details, if requested. It seems fair for people to expect those answers, if desired. Maybe that's just me. I was referring to a statement in the vein of "The strat should be >80% reliable to be considered reliable, not >50% reliable”. Where an arbitrary cutoff is can drastically influence analyses. I think this a consequence of the metric. Trying to go super fast in Lunatic is inevitably restrictive. This is intended, in order to resolve differences between characters. There are the optimal characters, then slightly less optimal ones, etcetc in a gradient until the characters are all ranked. I already noted that in this case it makes the ranking “optimal” > “utility” > “non-optimal in a gradient”, which some may see as unfortunate. This is a matter of tiering philosophy and the current criteria. It's something hard to resolve cleanly, but this way is at least self-consistent. Going slower results in more variable conclusions. Nowi is probably the case study for this. Why can we slow down enough for Nowi to shine but not slow enough for Donnel to shine? Or anyone, because everyone in this game is good when trained. I didn’t mean to imply that playlog is the only evidence we should consider. I just think it’s fairly compelling compared to other evidence we currently have for determining distinctions in Lunatic. It seems like it’s intuitive to go “somewhat briskly” and not whatever-who-cares pace. I suppose my point is then, why not just tend toward the minima of expected turn count (given some team composition)? I don’t think anyone anticipates calculating the exact values here all the time either. It merely sets up more rigorous framework for those tricky-to-judge cases that intuition cannot easily resolve. In Lunatic, I do not think a single turn (or 10) salvages Sully and Nowi. Re: Lon’qu: I think the point is, Lon’qu barely avoids the ohko. Other characters are still solidly 2hko’d or such. Axe users are the most common and hit the hardest. Avoiding all damage from them with high probability is a huge boon earlygame. I’d welcome evidence that Sully or someone can replicate his role. Re: Chrom. I originally thought he was too high, but his Spd Pair up is actually amazing for Frederick before Sumia, and then Lon’qu or Avatar or someone. It is extremely hard to double without +Spd. He apparently improves Endgame by a bunch, turncount and reliability wise (I’m theorizing Lucina can do it better, but we’ll ignore that for now) . I wouldn’t consider it the same way as Premonition because it is surprisingly complicated to get Chrom strong enough. It requires substantial planning throughout the game. It’s not really a interactive cutscene like Premonition. And there are alternate ways such as Braves+Ignis/Luna (requires more exp though). @Scarlet: If Seth was an armor knight, would FE8 be known as The Sacred Seth?

inb4 people start complaining about varying contexts. <_<

I think it's implied that lower expected turns are better. Maybe I'm misunderstanding your point. >_>

I don't like going as fast as possible because things like growths, accuracy, avoid don't matter. Bases and Move are completely overdominating. This is why I consider reliability a critical criteria. Redwall's formula is self-consistent (disclaimer: I didn't actually verify his math) and accounts for both turncount and reliability. There is no arbitrary limit here (except for the subjective value we, as players, assign being fast and reliable)